Thrombosis as a Treatment Complication in Hodgkin Lymphoma Patients

A Comprehensive Analysis of Three Prospective Randomized German Hodgkin Study Group (GHSG) Trials

S. Borchmann; H. Müller; I. Hude; M. Fuchs; P. Borchmann; A. Engert


Ann Oncol. 2019;30(8):1329-1334. 

In This Article



We analyzed thrombotic events in 5773 HL patients treated within the German Hodgkin Study Group's (GHSG) randomized clinical trials HD 13–15 between January 2003 and September 2009,[11,21,22] including early-favorable (HD 13), early-unfavorable (HD 14) and advanced-stage (HD 15) HL patients[2] (Table 1). Written informed consent, covering long-term follow-up within the database, was obtained from all patients. The study protocols were reviewed and approved by each participating institution's ethics committee. All trials were conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonization Guidelines for Good Clinical Practice. The trials were registered with under the following identifiers: ISRCTN63474366 (HD 13), ISRCTN04761296 (HD 14) and ISRCTN32443041 (HD 15).

Thrombotic Events

We report all arterial and venous thrombotic events within 1 year after trial enrollment, covering both treatment and early follow-up. We excluded incidental events and superficial vein thrombosis. Thrombotic events had to be confirmed by an appropriate imaging technique, such as Doppler ultrasound, computed tomography, magnetic resonance imaging or angiography. Events were first classified into venous and arterial thrombotic events. Next, we subclassified venous thrombotic events into deep vein thrombosis (DVT) of the arm, DVT of the leg, pulmonary embolism (PE) and other, rarer events (Table 2). Likewise, arterial events were subclassified into myocardial infarction, arterial thromboembolism of the leg and other, rarer events. In cases where multiple locations were identified in one patient, the combination was categorized as such. Patient characteristics and incidences are reported descriptively. The effects of disease stage and received treatment on thrombotic events were tested with logistic regression and Wald χ 2 statistics using SAS 9.4 and an alpha error probability <0.05 for significance.

Risk Factor Analysis

We studied a total of 23 potential risk factors (Table 3). Among those, single risk factors, such as a high platelet count, age or smoking status, and composite risk factor scores, such as the widely used Khorana score for the prediction of thrombosis risk in ambulatory cancer patients were considered.[12] As thrombotic events cumulated in advanced-stage HL, we focused on this subset of patients for our risk factor analysis and used the HD 15 data to compute multivariable logistic regression analyses to predict any or exclusively venous thrombotic events by treatment groups and, one by one, each other risk factor.